January 2019

Focus on Research: 5 Stages of Health Care Data Quality Maturity
By Vicki Mahn-DiNicola, RN, MS, CPHQ
For The Record
Vol. 31 No. 1 P. 28

Reaching a high level of trust in your organization's performance measurement data is dependent on people, processes, and technology all working together to ensure the highest levels of data quality possible. Data quality is crucial for properly measuring, managing, and improving your organization's clinical or financial performance ... and it's a lot like preparing a meal.

You see, the metrics that your organization monitors throughout the year are dependent on the quality of the data used to compile your results. Like a meal, the quality of the dish is going to depend on the quality of your ingredients. The fresher the ingredients, the greater chance you'll turn out a great dish that people will want to order again and again. Yet not all health care organizations are at the same stage of data quality maturity.

What kind of data ingredients does your hospital cook with? We break down the five stages of data quality maturity and provide a visual to help you identify where your organization is today.

Stage 1: Cooking From a Can
In the earliest stage of data quality maturity, the organization's quality department typically will look only at their data quality when problems arise. There are no reliable measures to routinely monitor the completeness or accuracy of the data, and there is little to no documentation of data definitions or standards.

While the organization may be able to fix data quality issues when they arise, they will most likely reoccur without formal processes in place to understand the root cause of their data decay.

Stage 2: Exploring the Basics
When a health care organization evolves into the second stage of data quality maturity, it is slightly more proactive about monitoring its data.

Data element standards and definitions are documented for commonly used terms, and there are basic measures in place to track incomplete or invalid data. Root cause analysis is conducted for simple data quality issues; however, it tends to be limited to the department level rather than an enterprisewide effort.

Stage 3: The Family Legend
In stage three of data quality maturity, the organization implements a more systematic process to measure, monitor, and manage its data quality. It has a comprehensive cookbook where all data definitions and standards are clearly defined and documented across the health care enterprise. There are technology solutions in place to help monitor and validate data integrity, along with defined processes for manual inspection when data decay occurs.

At this stage, root cause analysis is conducted at the enterprise level and basic data quality results are tracked and monitored within one or more applications within the organization.

Stage 4: Test Kitchen Entrepreneur
As an organization progresses to a higher level of data quality maturity, it has a well-established and well- managed process for change control, certification of data sources, and data exchange standards. Data quality is consistently measured, monitored and managed, allowing the organization to be highly proactive rather than reactive. Just like a test kitchen for new recipes, these organizations have robust test and reference data sets to evaluate the impact of changing data sources on their measure outputs. When the soup gets salty, there are well-managed people, processes, and technologies in place to conduct root cause analysis to prevent future data decay. Data quality performance is shared with the measure owners and data stewards, and data quality is an essential part of the quality management culture.

Stage 5: Michelin Three-Star Restaurant
At the highest level of data quality maturity, the organization is functioning like a Michelin three-star kitchen. It has well-governed and broadly adopted policies and procedures in place for managing performance data and data quality across the entire health care enterprise.

Technology platforms support fully automated data quality surveillance and dashboards, which include a metadata warehouse to manage the data that describe the data. Detailed data elements describe how each data element is used in the organization, with rigorous change control processes in place when IT systems are updated with data conversions or when new applications are implemented.

Data quality measurements, minimum thresholds, and benchmarks are communicated to both measure owners and end user stakeholders. Stage 5 organizations operate proactively and achieve the highest levels of trusted performance data possible in the world of ever-changing measure specifications, data taxonomies, and IT environments.

— Vicky Mahn-DiNicola, RN, MS, CPHQ, is the vice president of clinical analytics and research at Medisolv.

To view the entire report, visit www.medisolv.com.